INDEX

Explanations

contractions and forms of "to have" indicating past experiences or states

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ROY

-0.14

rex

-0.14

yll

-0.14

roy

-0.13

yz

-0.13

.cbo

-0.13

AWN

-0.13

aki

-0.13

-bound

-0.13

POSITIVE LOGITS

 previously

0.19

ipple

0.17

 Previously

0.17

Ìī

0.16

plx

0.16

imens

0.16

Previously

0.15

reno

0.15

ÑĢÐµÑģ

0.15

CHASE

0.15

Activations Density 0.289%

contractions and forms of "to have" indicating past experiences or states

No Comments

No Known Activations